100 Best Open-Source Agent Frameworks in 2026 (Ranked by GitHub Activity)

DeepSeek Harness: Everything is a Plugin.

The agent that grows with you

OpenHuman is the fastest, cheapest, most efficient open-source agent harness. Written in Rust

LLM inference in C/C++

OmO: Just type "mass ulw" keyword with your prompt. Now you are the master of graph engineering.

Open-source LLM engineering platform for observability, evaluation, prompt and dataset management

Open-source platform for building, scheduling, and managing collaborative AI agent teams

AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs

The agent engineering platform

A natural language interface for computers

Transforms complex documents like PDFs into LLM-ready markdown/JSON for your Agentic workflows.

Production-ready platform for agentic workflow development.

🙌 OpenHands: AI-Driven Development

"Vibe-Trading: Your Personal Trading Agent"

Teams-first Multi-agent orchestration for Claude Code

Hermes WebUI: The best way to use Hermes Agent from the web or from your phone!

AI Agent Framework, the Pydantic way

Low-code multi-agent AI framework for planning, research, coding, and cross-platform delivery

DeepTutor: Lifelong Personalized Tutoring. https://deeptutor.info/.

Open-source meta-harness for orchestrating Claude Code, Codex, Cursor, Pi, and custom AI agents

Agent framework for multi-agent swarms, autonomous workflows, adaptive memory, and vector RAG

The AI Toolkit for TypeScript. From the creators of Next.js, the AI SDK is a free open-source library for building AI-powered applications and agents

Agentic coding tool that understands codebases and handles tasks and Git workflows from your terminal

Run and supervise teams of coding agents from planning to merge. Any harness (Claude code, codex, +25 more). Desktop, web, mobile, and cloud agents.

Coordinated engineering system for AI coding agents with planning, testing, review, memory, and security

Get your documents ready for gen AI

Get up and running with Kimi-K2.5, GLM-5, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.

Debug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.

🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.

Local Claude Code framework for evaluating jobs, tailoring CVs, writing cover letters, and interview prep

Build, run, and manage agent platforms.

Temporal service

An Open Source implementation of Notebook LM with more flexibility and features

Use Claude Code as the foundation for coding infrastructure, allowing you to decide how to interact with the model while enjoying updates from Anthropic.

Build resilient language agents as graphs.

Brigade — Your personal intelligence, built enterprise-grade

TradingAgents: Multi-Agents LLM Financial Trading Framework

LocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.

Cross-platform desktop workspace for Claude Code with multi-agent sessions, Git worktrees, and diff review

🚀 PR Agent - The Original Open-Source PR Reviewer. This repo is not the Qodo free tier! Try the free version on our website.

Open-source AI assistant framework for planning tasks, running tools and skills, memory, and multi-agent teams

Open-source agent chatbot platform for building AI applications in instant messaging workflows

Build, run, manage agentic software at scale.

Fabric is an open-source framework for augmenting humans using AI. It provides a modular system for solving specific problems using a crowdsourced set of AI prompts that can be used anywhere.

Open-source Python framework for self-hosted AI agents with WebUI, tools, memory, MCP, and automation

The most RAM efficient harness

A framework for building, orchestrating and deploying AI agents and multi-agent workflows with support for Python and .NET.

Jan is an open source alternative to ChatGPT that runs 100% offline on your computer.

Powerful AI Client

Open source agentic operating system

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

Your Personal AI Assistant; easy to install, deploy on your own machine or on the cloud; supports multiple chat apps with easily extensible capabilities.

Trigger.dev – build and deploy durable AI agents and workflows

Ekko Studio is a local-first AI workspace for multi-agent chat, coding, and visual workflows, available on desktop and the web.

DSPy: The framework for programming—not prompting—language models

AI coworker with memory and collaboration

Fully autonomous AI Agents system capable of performing complex penetration testing tasks

LLM-powered multi-market stock analysis with multi-source data, decision dashboards, alerts, and notifications

Markdown skills for autonomous ML research: cross-model review, idea discovery, experiment automation

Free, cross-platform database client and SQL workspace with AI assistance for 40+ databases

Patterns, starters, and CLI tools for orchestrating AI coding agents, verifying results, and persisting state

An autonomous agent that conducts deep research on any data using any LLM providers

The official Python SDK for Model Context Protocol servers and clients

Open-source AI video creation platform with an infinite canvas, agents, and visual workflows

Story Creation AI Agent for novel, scripts, translation, interactive games, and IP content

AI Agent Engineering Platform built on an Open Source TypeScript AI Agent Framework

A collection of specialized AI agents with distinct expertise, personalities, and workflows

Open-source, self-hosted AI trading platform for Python strategies, backtesting, and paper or live trading

Model Context Protocol Servers

The official TypeScript SDK for Model Context Protocol servers and clients

The ultimate LLM/AI application development framework in Go.

Open-source platform for building AI-powered IM bots with LLMs, tool calling, and RAG knowledge bases

Use Garry Tan's exact Claude Code setup: 15 opinionated tools that serve as CEO, Designer, Eng Manager, Release Manager, Doc Engineer, and QA

Data validation using Python type hints

Open source software that helps you create and deploy high-frequency crypto trading bots

The open source AI research agent.

Access large language models from the command-line

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

UFO³: Weaving the Digital Agent Galaxy

AI assistant running agents in Linux containers with messaging, memory, and scheduled jobs

Framework to build resilient language agents as graphs.

GUI Agent for E2E Testing

Chinese community edition with 277 AI agent personas across 20 departments and multi-agent orchestration

Public repository for Agent Skills

Turn Claude Code into a full game dev studio — 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.

Agent2Agent (A2A) is an open protocol enabling communication and interoperability between opaque agentic applications.

Go ahead and axolotl questions

NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

Open-source agentic video production system for research, scripting, asset generation, and editing

🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org

Free, Unlimited AI Code Reviews That Run on Commit

🤗 smolagents: a barebones library for agents that think in code.

open-source agentic AI data assistant for the next generation of AI + Data products.

Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1

Create state-machine-powered LLM agents using XState

Reliable Multi-Agent Orchestration Framework

"DeepCode: Open Agentic Coding (Agent Harness & Loop Engineering & Multi-Agent Orchestration)"

A generative AI extension for JupyterLab